diff --git a/doc/amex-file-format.org b/doc/amex-file-format.org deleted file mode 100644 --- a/doc/amex-file-format.org +++ /dev/null @@ -1,66 +0,0 @@ -Amex CSV file format - -Current as of December 2, 2012 - -This documents the Amex CSV that can be downloaded from the Amex -website. Obtain this by using the "Download" link that is available -when viewing account transactions. Be sure to select the checkbox for -"additional transaction details" or the like, as this format contains -the unique Amex ID numbers. - -Newlines are with linefeeds only (UNIX style, not Windows style.) -Although newlines separate records, they also appear within some of -the fields, so you cannot split the file into records using newlines -alone (which makes it difficult to use line-oriented tools like awk.) - -Here are the fields, in order. The fields are separated by commas. The -list starts at 1 (org-mode does not seem to like starting lists at 0.) - -Many fields are unknown. These have always appeared blank in the data -I have. - -1. Date. This is given as MM/DD/YYYY followed by a three-digit day of - the week, such as "Wed". - -2. Unknown - -3. Description. This is the payee name (often truncated) followed by - the city and state. This contains many spaces (presumably so it - would line up in a table?) Appears wrapped in double quotes. - -4. Unknown - -5. Unknown - -6. Unknown - -7. Unknown - -8. Amount. Charges appear simply with the amount; payments and credits - appear with a prepended minus sign. This field is not wrapped in - double quotes. - -9. Unknown - -10. Category. Appears to categorize the merchant (e.g. "Hardware - Store") but the exact format is not immediately apparent; - sometimes a number appears with the description, sometimes there - are just numbers. This field is wrapped in double quotes. - -11. Doing Business As. This seems to be the best place to pull the - payee information from, as the Description field (number 3) mixes - this up with other information. Double-quoted. - -12. Address. Contains a street address with a city and state. Looks - something like Street Address (newline) City (newline) State - 2-digit. Note the embedded newlines. Double-quoted. - -13. Postcode. Postal code, then newline, then country. Note the - embedded newline. Double-quoted. - -14. Amex ID. A long integer. Appears within both double quotes and - single quotes, e.g. "'320123300342679128'". - -15. Unknown - -16. Unknown diff --git a/doc/amex-workflow.org b/doc/amex-workflow.org deleted file mode 100644 --- a/doc/amex-workflow.org +++ /dev/null @@ -1,68 +0,0 @@ -Amex Workflow - -Here is a proposed workflow for dealing with the download of Amex -transactions and the reconciliation of Amex statements. - -* Import to database using amex-import - -This will parse downloaded data from the Amex website and import them -to the database of all Amex transactions and will assign all -transactions a "U-number." The U-number is a unique number that -identifies an Amex transaction. U-numbers are used because, although -Amex assigns unique identifiers to transactions, they are extremely -long. - -* Merge from database to ledger file using amex-merge - -Checks to ensure that every transaction that is in the Amex database -is represented in the ledger. Follows these steps: - -- Is the database transaction's U-number found in the numbers for the - Amex transactions that are in the ledger? If so, stop. - -- Otherwise, go through existing Amex ledger transactions that do not - have a U-number and see if any of these match the database - transaction. There is a match if all of the following are true: - - - The quantities match exactly; - - - The ledger transaction is a debit or a credit, as appropriate (if - Amex shows an increase in the balance, then the ledger shows a - credit; for payments, the ledger shows a debit.) - - - The date on the Amex transaction matches the local date on the - ledger transaction. - -- If no matches are found, create a transaction and append it to the - end of the list of Ledger transactions. Assign the appropriate - U-number. Prints new ledger file to stdout. - -After running amex-merge, study the results. If necessary, assign the -U-numbers to already existing ledger transactions, or edit the newly -created transactions. - -* Clear using amex-clear - -This command will also work with a list of downloaded transactions. It -will follow these steps: - -- Make sure that all the downloaded transactions are in the - database. If not, quit with an error. - -- Make sure that all the downloaded transaction U-numbers are also in - the list of ledger transactions. If not, quit with an error. - -- Make sure that each ledger posting corresponding to the downloaded - transaction has no flag for the posting itself (a flag for the - transaction is OK.) It one has a flag, quit with an error. - -- Assign the flag "C" to each transaction. Prints new ledger file to stdout. - -After running amex-clear, make sure that the balance of all cleared -and reconciled ledger Amex postings is the same as the Amex statement -balance. - -* Reconcile using amex-reconcile - -This command simply changes the "C" flags in the Amex ledger account -to "R" flags. diff --git a/doc/bofa-file-format.org b/doc/bofa-file-format.org deleted file mode 100644 --- a/doc/bofa-file-format.org +++ /dev/null @@ -1,159 +0,0 @@ -Bank of America download file format - -This is the format for the "WEB Connect for Quicken 2010 and above" -option that is on the Bank of America download format. Ostensibly this -is OFX in SGML, but I don't know if this is actually true as I have -not yet found much about what "WEB Connect" is. A simple CSV would be -preferable; however, B of A does not make unique transaction IDs -available in CSV format. (QIF does not have this information, neither -does the "Microsoft Excel format" on the website, which appears to -just be a CSV.) - -The file has DOS line endings (\r\n). - -Unlike XML, tags are not always closed. The tag names are always in -all caps. It does appear that tags that open alone on a line are -balanced, while tags that are followed on the same line with data are -not balanced. - -* Header - -There is a header portion with nine lines. Each has the field name, -followed by a colon, and then some information (no space between the -colon and the data). They have field names as follows: - -- OFXHEADER -- DATA -- VERSION -- SECURITY -- ENCODING -- CHARSET -- COMPRESSION -- OLDFILEUID -- NEWFILEUID - -Then there is a blank line. - -* OFX tag -This tag is balanced. It encloses all the data. All other tags are -within this tag. - -* SIGNONMSGSRSV1 -Sign on message? Balanced. Encloses some tags: - -** SONRS -Balanced. - -*** STATUS -Balanced. - -**** CODE -Not balanced. There is a digit on the same line. - -**** SEVERITY -Not balanced. Says "INFO" on the same line. - -*** DTSERVER -Not balanced. Has some sort of date-time information? - -*** LANGUAGE -Not balanced. Says ENG. - -*** DTACCTUP -Not balanced. Has some date-time information. Looks like same -information in DTSERVER. - -*** FI -Balanced. Financial institution? - -**** ORG -Not balanced. Bank of America. - -**** FID -Not balanced. Says 5959. - -*** INTU.BID -Not balanced. Four-digit number. - -*** INTU.USERID -Not balanced. Has your Bank of America user ID. - -* BANKMSGSRSV1 -Balanced. - -** STMTTRNRS -Balanced. - -*** TRNUID -Not balanced. "0" - -*** STATUS -Balanced. - -**** CODE -Not balanced. "0" - -**** SEVERITY -Not balanced. "INFO" - -*** STMTRS -Balanced. - -**** CURDEF -Not balanced. "USD" - -**** BANKACCTFROM -Balanced. - -***** BANKID -Not Balanced. Digits. - -***** ACCTID -Not balanced. Digits. - -***** ACCTTYPE -Not balanced. Checking - -**** BANKTRANLIST -Balanced. - -***** DTSTART -Not balanced. Digits. Some sort of date-time. - -***** DTEND -Not balanced. Digits. Some sort of date-time. - -***** STMTTRN -Balanced. Where interesting stuff starts. - -****** TRNTYPE -Not balanced. Transaction Type? Mostly CREDIT or DEBIT in the data I -have, though it might also be ATM or who knows what else. - -****** DTPOSTED -Not balanced. A date-time. I am guessing the format for a date-time is -YYYYMMDDHHMMSS. B of A is leaving the MMSS as zeroes. - -****** TRNAMT -Not balanced. Digits with a decimal point. CREDIT transactions have no -minus sign; DEBIT transactions have a leading minus sign. - -****** FITID -Not balanced. Financial Institution ID? Digits with some points and -maybe a minus. - -****** NAME -Not balanced. The payee name. - -****** CHECKNUM -Appears on checks. Not balanced. - -**** LEDGERBAL -Balanced. - -***** BALAMT -Not balanced. An amount. - -***** DTASOF -Not balanced. Date as of? A date-time. Unlike the others, this -date-time has minutes and seconds. diff --git a/doc/penny-ofx.ini b/doc/penny-ofx.ini deleted file mode 100644 --- a/doc/penny-ofx.ini +++ /dev/null @@ -1,70 +0,0 @@ -# Sample configuration file for penny-ofx - -# Optional default financial institution account. If you do not -# specify a default account, penny-ofx will require that you specify -# one using the -f option every time you run penny-ofx. -default_fit_acct = bigbank - -# Example: a credit card from Bigbank. All of the fields shown are -# required. You can configure multiple financial institution accounts. -[bigbank] - -# File in which to store database information for this account. You -# must use a different file for each account. Therefore, be careful if -# you cut and paste when you create new accounts in this file, as you -# must change the filename here. -file = /home/massysett/ledger/bigbank.bin - -# Postings from this account appear under this account in your ledger -# file(s). -penny_acct = Liabilities:Current:Bigbank - -# When penny-fit finds a financial institution posting and it -# does not have a matching posting in your ledger, it must create -# a new transaction with two postings. One posting will be in the -# pennyAcct specified above and the other posting will be in this -# account. -default_acct = Expenses:Unclassified - -# All postings will be in this currency -currency = $ - -# The next two options control digit grouping when penny-ofx reprints -# your ledger. groupLeft and groupRight control whether digits are -# grouped to the left and to the right of the radix point, -# respectively. Available choices are: -# -# none - do not group any digits -# -# large - group only if the number to be grouped is greater than 9,999 -# (if grouping to the left of the decimal point) or if there are more -# than 4 decimal places (if grouping to the right of the decimal -# point.) -# -# all - group whenever there are at least four digits to be grouped. -group_left = all -group_right = none - -# Postings from your financial institution are specified in terms of -# increases or decreases. Postings in your ledger are specified in -# terms of debits or credits. The translator specifies how to convert -# a posting from your financial institution to a posting in the -# pennyAcct in your ledger. -increase_is = Credit - -# When penny-fit creates new postings it must put the commodity -# either to the left of the quantity or to the right of the -# quantity. -commodity_on = Left - -# When penny-fit creates new postings it must decide whether to -# put a space between the commodity and the quantity. -space_between = False - -# Give some helpful information here about where to find the file on -# your bank website. You can use multiline strings, as shown, by -# indenting subsequent lines. -more_info = Statements from bigbank. - Bigbank makes its statements available by clicking on - "Account Info", then logging in, then clicking on - "Statements". diff --git a/install-docs b/install-docs --- a/install-docs +++ b/install-docs @@ -20,7 +20,7 @@ done install -m755 -d $DOCDIR - for filename in doc/*.{org,dot}; do + for filename in doc/*.{org,dot,hs}; do install -v -m644 $filename $DOCDIR done diff --git a/man/penny-fit.7 b/man/penny-fit.7 --- a/man/penny-fit.7 +++ b/man/penny-fit.7 @@ -188,6 +188,7 @@ decrease:IncreaseIsCredit:credit:No .TE . +.P If .B penny-fit finds a match for a financial institution posting in this way, then it diff --git a/man/penny-reconcile.1 b/man/penny-reconcile.1 --- a/man/penny-reconcile.1 +++ b/man/penny-reconcile.1 @@ -14,7 +14,10 @@ .IR C . Changes those flags to the letter .I R -and prints the resulting ledger to standard output. This is useful +and prints the resulting ledger to standard output. +. +.P +This is useful when reconciling a financial institution account: you may first mark postings in your ledger that match a posting on your bank statement with a @@ -32,6 +35,7 @@ .B penny-reconcile will automatically mark all the posts reconciled. . +.P If no .IR FILE ", or " FILE " is " - , read standard input. @@ -72,6 +76,7 @@ This would happen only rarely though, so you might be just fine using a query-replace function in your text editor. . +.P Also, .B penny-reconcile will tidy up your ledger file--that is, it might rearrange or delete diff --git a/penny-bin.cabal b/penny-bin.cabal --- a/penny-bin.cabal +++ b/penny-bin.cabal @@ -1,5 +1,5 @@ Name: penny-bin -Version: 0.14.0.0 +Version: 0.16.0.2 Cabal-version: >=1.8 Build-Type: Simple License: BSD3 @@ -173,7 +173,6 @@ , README , doc/*.org , doc/*.dot - , doc/*.ini , doc/*.hs , examples/*.pny , man/*.1 @@ -187,7 +186,7 @@ Executable penny Build-depends: base ==4.*, - penny-lib ==0.14.* + penny-lib ==0.16.* Main-is: penny-main.hs Other-modules: Paths_penny_bin @@ -201,7 +200,7 @@ Executable penny-selloff Build-depends: base == 4.*, - penny-lib ==0.14.*, + penny-lib ==0.16.*, explicit-exception ==0.1.*, containers ==0.5.*, semigroups ==0.9.*, @@ -221,7 +220,7 @@ Executable penny-diff Build-depends: base ==4.*, - penny-lib ==0.14.*, + penny-lib ==0.16.*, text ==0.11.*, multiarg ==0.16.*, explicit-exception == 0.1.* @@ -239,7 +238,7 @@ Build-depends: base ==4.* , multiarg ==0.16.* - , penny-lib ==0.14.* + , penny-lib ==0.16.* , pretty-show ==1.5.* , text ==0.11.* @@ -252,7 +251,7 @@ Executable penny-reconcile Build-depends: base ==4.* - , penny-lib ==0.14.* + , penny-lib ==0.16.* , text ==0.11.* , multiarg ==0.16.* , explicit-exception ==0.1.* diff --git a/penny-main.hs b/penny-main.hs --- a/penny-main.hs +++ b/penny-main.hs @@ -238,16 +238,16 @@ lightEvenTextSpec = id lightOddTextSpec :: Chunk -> Chunk -lightOddTextSpec = id .+. color8_b_default .+. color256_b_255 +lightOddTextSpec = (<> (c8_b_default <> c256_b_255)) lightDebit :: (Chunk -> Chunk) -> Chunk -> Chunk -lightDebit f = f .+. color8_f_magenta .+. color256_f_52 +lightDebit f c = f c <> c8_f_magenta <> c256_f_52 lightCredit :: (Chunk -> Chunk) -> Chunk -> Chunk -lightCredit f = f .+. color8_f_cyan .+. color256_f_21 +lightCredit f c = f c <> c8_f_cyan <> c256_f_21 lightZero :: (Chunk -> Chunk) -> Chunk -> Chunk -lightZero f = f .+. color8_f_black .+. color256_f_0 +lightZero f c = f c <> c8_f_black <> c256_f_0 -- | The dark color scheme. You can change various values below to -- affect the color scheme. @@ -271,16 +271,16 @@ darkEvenTextSpec = id darkOddTextSpec :: Chunk -> Chunk -darkOddTextSpec = id .+. color8_b_default .+. color256_b_235 +darkOddTextSpec = (<> (c8_b_default <> c256_b_235)) darkDebit :: (Chunk -> Chunk) -> Chunk -> Chunk -darkDebit f = f .+. color8_f_magenta .+. color256_f_208 +darkDebit f c = f c <> c8_f_magenta <> c256_f_208 darkCredit :: (Chunk -> Chunk) -> Chunk -> Chunk -darkCredit f = f .+. color8_f_cyan .+. color256_f_45 +darkCredit f c = f c <> c8_f_cyan <> c256_f_45 darkZero :: (Chunk -> Chunk) -> Chunk -> Chunk -darkZero f = f .+. color8_f_white .+. color256_f_15 +darkZero f c = f c <> c8_f_white <> c256_f_15 -- | Plain scheme has no colors at all.